NERC training course: Quantifying The Environment II

The second 1 week NERC sponsored training course on Quantifying the environment will be run from 5-9 Jan 2015 at the University of Glasgow. This follows on from our previous highly successful one week courses. NERC-funded PhD students and EC Researchers will be sponsored covering travel, subsistence and course registration.

The aim of this short course is to provide environmental scientists with a thorough knowledge of and training in three key areas of statistical development- flexible regression methods, spatio-temporal modelling and functional data analysis. The course includes practical sessions using R.

Registration will open on 5th October and will close on Nov 30th.

The course will start at Mon lunchtime, with flexible regression methods with an introduction to some of the theory and application of advanced regression models including, nonparametric and generalised additive models, and quantile regression in environmental contexts. A variety of approaches for smoothing will be explained and illustrated through a series of lectures and practical lab sessions.  The sessions and lectures will illustrate the appropriate uses and restrictions of advanced regression models, using R.

This will then be followed by Spatio-temporal modelling, covering an introduction to statistical approaches to modelling data that have spatial and temporal structure. The sessions will first give in-depth discussions of purely spatial and purely temporal modelling, including: geostatistics (including Kriging), areal (lattice) models including Markov Random fields and point process models including homogeneous and inhomogeneous Poisson processes. An introduction to spatio-temporal modelling, dealing first with separable spatial and temporal correlation structures, before finally addressing a full spatio-temporal construction will conclude this topic.

The final sessions will be on Functional data analysis which is a new and very powerful statistical methodology, which treats time series data in new ways (the "datapoint" becomes the curve). These sessions will start with an introduction to methods in functional data analysis, with an emphasis on practical issues and data arising from environmental monitoring devices and optical or mechanical tracking devices. The sessions will train students to identify scenarios where data may be considered to be smooth functions and construct visualization strategies and implement nonparametric smoothing for exploring functional data. Using several environmental data sets we will illustrate ways to describe the variation among a group of curves, to describe differences between groups of curves and to understand the effect of one set of curves on another by formulating and fitting several types of functional linear models.

Further details will be announced at: https://www.gla.ac.uk/schools/mathematicsstatistics/events/details/?id=9365